Inferring Relationships Between Indigenous Baja California Sur and Seri/Comcáac Populations Through Cultural Traits Author(S): SHANE J

Home , Francisco de Ulloa, Oklahoma, Sierra de la Giganta

Inferring Relationships Between Indigenous Baja California Sur and Seri/Comcáac Populations Through Cultural Traits Author(s): SHANE J. MACFARLAN and CELESTE N. HENRICKSON Source: Journal of California and Great Basin Anthropology, Vol. 30, No. 1 (2010), pp. 51- 67 Published by: Malki Museum, Inc. Stable URL: https://www.jstor.org/stable/23215635 Accessed: 11-01-2019 05:39 UTC

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at https://about.jstor.org/terms

Malki Museum, Inc. is collaborating with JSTOR to digitize, preserve and extend access to Journal of California and Great Basin Anthropology

This content downloaded from 142.103.160.110 on Fri, 11 Jan 2019 05:39:20 UTC All use subject to https://about.jstor.org/terms Journal of California and Great Basin Anthropology [ Vol. 30, No. 1 (2010) | pp. 51-67

Inferring Relationships Between Indigenous Baja California Sur and Seri/Comcáac Populations Through Cultural Traits

SHANE J. MACFARLAN Washington State University, Department of AnthropologyVMMC 202,14204 NE Salmon Creek Ave.,Vancouver,WA 98686 [email protected]

CELESTE Ν. HENRICKSON University of California, Berkeley, Department of Anthropology, 2251 College Building, Berkeley, CA 94720-9637 [email protected]

Cultural relationships between historic Baja California Sur indigenous groups are poorly understood, despite the presence of historical accounts of these peoples. Relationships between groups largely have been reconstructed through linguistic attributes recorded by Jesuit missionaries. Non-linguistic cultural traits derived from historical accounts also can be used to determine relationships between groups. We use content analysis to systematically organize cultural trait information from historical explorer accounts, Jesuit missionary documents, and academic research pertaining to three Lower California groups: the Pericú, Guaycura, and southern Cochimi, as well as the Seri/Comcáac Indians of the Gulf of California and Sonora. Reliability analyses reveal considerable cultural homogeneity between the southern Cochimi and Guaycura, and cultural dissimilarity between all other groupings. Hierarchical cluster analysis reveals the Guaycura and southern Cochimi comprise a distant, but natural grouping with the Seri/Comcáac, while the Pericú are isolated. Several hypotheses are presented to explain these patterns.

In general, simplicity and crudity of workmanship increases from north to south, attaining in Southern and Lower California the greatest degree of primitiveness, a condition which was undoubtedly due to the area being marginal to its ancestral hearth and farthest removed for many centuries from the points of ingress of foreign influence [Rogers 1945:168].

If we are too quick to assume marginality from a reading of privileged documentary sources, or an emotional response to places which to us seem remote.. .we are in danger of destroying the history and lives of the people we purport to study [Horning 2007:374],

peninsular geography constrains emigration and the THE is deeplyCONCEPT rooted OF in PENINSULARthe study of Baja ISOLATION consequentCalifornia flow of peoples into and within the region. prehistory. The cultural geography of the Baja Indeed, California the geographic ruggedness of Baja California, peninsula is often described using terms whichlike "cul-de hampers access to many areas today, has facilitated sac" (Massey 1966), "population trap" (Tuohy its characterization 1978), as culturally marginal both in the or "refuge" (Gonzalez-José et al. 2003). These past phrasesand within modern research frameworks of North imply a cultural core rooted near the northern America base (e.g., Aschmann 1967; Dalton 2005; Gonzalez of the peninsula and an increasingly marginalized José et al. or2003; Massey 1966; Rogers 1945). Much of the isolated periphery moving toward the southern discussion tip. about marginality is reinforced by geographic This imagery can influence our interpretations patterns of involving the the distribution of cultural traits in the past (Horning 2007) and lead to simplified south concepts of the peninsula—for example, the retention of of peninsular history, including the manner atlatls in (Laylanderwhich 2007), distinctive cave burials (Massey

1955), and a notable absence of indigenous pottery Several factors contribute to this situation, including (1) (Massey 1966). Imagery of Baja California Sur (hereafter the absence of widespread, well-dated archaeological referred to as BCS) cultural isolation also is exacerbated sites; (2) disagreements between ethnohistorical accounts by missionary accounts that expressed feelings of isolation, regarding the number of linguistic groups present and fear, and the remoteness of living so far from mainland their associated territories (e.g., del Barco 1981; Massey Mexico and Europe (e.g., Baegert 1952). A superficial 1949); and (3) ethnohistorical accounts that confound reading of these accounts could allow one to conflate linguistic with group affiliation (Massey 1949). Also, like missionary experiences of isolation with the historical many regions in North America, it is difficult to correlate cultural relationships between indigenous groups. The prehistoric sites with historic indigenous populations. extent to which a region or people are marginal or These problems have been compounded by the effects isolated is an academic matter that requires empirical of disease vectors (Aschmann 1967; Cook 1937), forced testing through inferential statistics. As a peninsula of missionary relocation programs (Jackson 1983,1984), culturally heterogeneous groups, an important first step and the possibility of historical demographic shifts and towards positioning Baja California within a wider North the geographic displacement of indigenous populations American prehistory is to understand the relationships due to indigenous agency (Mathes 1975) — although it between the BCS groups themselves and with their should be noted that the latter factor has been called neighbors. into question (cf. Laylander 1997:16). A final variable Missionary accounts and archaeological data suggest confounding the interpretation of relationships between that considerable cultural heterogeneity existed between historic BCS indigenous populations involves the immediate groups within BCS (Laylander 2000). question of whether one examines linguistic or material From north to south, the groups discussed here are the culture attributes. southern Cochimi, Guaycura, and Pericú (Fig. l).To date, Missionary letters and reports are the most archaeological excavations and historic accounts indicate productive source of information for illuminating these groups were small-scale, mobile foragers who cultural relationships between Baja Californian groups existed within diverse ecosystems that traversed extreme (Aschmann 1986). These historical documents provide gradients from coast to coast (Ashmann 1967; Hugo and demographic, linguistic, and cultural trait information Exequiel 2007; Hyland 1997; Ritter 2006). At the time of pertaining to specific indigenous groups. Of these, contact, Pericú territory included the peninsula's southern linguistic attributes are overwhelmingly preferred cape, as well as the four islands of Espíritu Santo, La for reconstructing prehistoric relationships; however, Partida, San José, and Cerralvo. Directly to the north of multiple interpretations exist as to the appropriate the tropical cape were the Guaycura, who inhabited the clustering of historic groups. Massey (1949, 1966) southern Sierra de la Giganta and the Magdalena coastal proposed the presence of two language families (Yuman plain. To the north of the Guaycura were the linguistically and Guaicurian) subdivided into four linguistic groups diverse Cochimian groups. Additionally, the Seri are (the Peninsular Group or Cochimi, Guaicura, Huchiti, included here, a group located beyond the peninsula and Pericú). Gursky (1966) and Swadesh (1967) suggested on the west coast of mainland Mexico, who currently that Guaycura was best placed within the Hokan refer to themselves as the Comcáac (Bowen 2000). Oral language stock, while Fernandez de Miranda (1967) and histories (Bowen 1976,2000) and archaeological research Campbell (1997) disagreed. Kroeber (1931) considered (Bowen 1976; Foster 1984) suggest the Seri/Comcáac had Cochimi to be related to Yuman, but did not afford it cultural contacts during the historic period with peoples full membership status, while Troike (1976) and Mixco from the Baja California peninsula. (1978, 2006) suggested that Cochimi and Yuman were Relative to other regions in North America, genetically related but should be considered two distinct sociocultural interactions among historic, indigenous families. Mixco (2006) has stated that the Guaycuran and groups in BCS are not well understood, despite the Cochimian languages are probably unrelated. Regardless existence of a variety of ethnohistorical accounts of of how the linguistic relationships are characterized, the region dating from the sixteenth century onward. a purely linguistic appraisal by no means provides

This content downloaded from 142.103.160.110 on Fri, 11 Jan 2019 05:39:20 UTC All use subject to https://about.jstor.org/terms ARTICLE I Inferring Relationships Between Indigenous Baja California Sur and Seri/Comáac Populations Through Cultural Traits | Maofarlan / Henrlckson

Figure 1. Cultural groups dicussed in the text..

definitive answers to the cultural relationships between that the Pericú and Cochimí were culturally dissimilar, populations, given that linguistically distinct groups while the Guaycura shared cultural features with both, often have considerable cultural contact (e.g., as in based on the distribution of a limited number of such northern California — see Jordan and Shennan 2003; cultural elements as the presence/absence of reed O'Neil 2008). The same documents used to reconstruct boats, atlatls, fishing technology, and basketry. Although linguistic relationships also can be used to reconstruct Massey was an expert on the ethnohistoric record cultural relationships between groups, if one focuses and peoples of Baja California, he did not formally the analysis on cultural traits — that is, on institutional analyze the distribution of cultural trait data to test structures (e.g., marriage patterns, residence rules) that these relationships. Therefore, his suggestions should be organize societies or ethnic markers (arbitrary, visual considered the opinions of an expert and propositions expressions of group membership that are thought to worthy of further empirical testing. facilitate within-group social action) (Brown 2008). Kroeber (1931) summarized the distribution of According to McElreath, Boyd, and Richerson (2003), cultural traits among the Guaycura, Cochimi, and to ethnic markers such as adornment, dress, or hairstyle a lesser extent the Pericú (as well as several other allow people to identify those individuals who share a non-peninsular groups), utilizing the Jesuit missionary common underlying normative framework for behavior, accounts of Johann Jacob Baegert, as well as research which facilitates mutually beneficial social interactions by Miguel Venegas and Francisco Xavier Clavijero while simultaneously excluding outsiders. (1789), in order to determine their relationships with Treating historical documents as ethnographies can the Seri/Comcáac. Kroeber tentatively suggested that be problematic (Mathes 1981); however, these documents (1) the Guaicura had a greater cultural affinity to are the only firsthand accounts of historical groups from the Seri/Comcáac than did the Cochimi; and (2) the BCS. Missionaries and explorers lived amongst these cultural groups of lower California as a whole were indigenous groups for periods ranging from several days less culturally similar to the Seri/Comcáac than the to upwards of 30 years. Combined, their reports represent Gila Pima or Walapai-Havasupai. His analysis involved over 250 years of direct observations of past ecosystems, the addition of the number of traits each group shared landscapes, and cultural behavior. The record, at times, with the Seri/Comcáac subtracted from the number describes in detail the cultural inventories of several of traits they did not share; however, he did not use distinct social and/or linguistic groups, such as the Pericú, the same cultural trait distributions to measure all Guaycura, Cora, and Cochimi (e.g., Baegert 1952; relationships. del Instead, the number and kind of traits Barco 1981; Napoli 1970). Included in this inventory areused to determine relationships differed for each pair ritual practices, watercraft technology, and the clothing wise grouping with the Seri/Comcáac. As a result, there styles and toolkits of men and women (for reviews, issee no way to meaningfully interpret the nature and Laylander 2000 or Mathes 2006). Archaeologists andmagnitude of the differences between groups, as each ethnographers use this information to contextualize was measured on a separate metric. When inferences of sites and case studies (e.g., Aschmann 1967,1986; Bowen group similarity are based on the presence or absence 1976, 2000; Kroeber 1931; Massey 1966). Recently, of cultural traits, and information on those cultural traits archaeologists (e.g., Laylander 1997,2000) and historians is missing for some groups, both Type I (incorrectly (Mathes 2006) have compiled these cultural inventories concluding that a difference exists between groups when into several fine works of scholarship (and to which nonethe is present) and Type II (incorrectly concluding authors are greatly indebted). that a similarity exists between groups when in fact they Although cultural traits have been used in the past are different) errors are inflated, causing a distorted for reconstructing cultural relationships within pictureBaja of relationships to emerge. Although it was a California (e.g., Massey 1947,1966), and between valiantBaja first attempt, the paucity of data and the lack California and other cultural regions of North America of statistical controls render any interpretation difficult. (e.g., Kroeber 1931), none has employed inferential Bowen (1976:102) echoed similar concerns about the statistics. Massey (1947:346), for example, suggested interpretation of Kroeber's data, but suggested that

This content downloaded from 142.103.160.110 on Fri, 11 Jan 2019 05:39:20 UTC All use subject to https://about.jstor.org/terms ARTICLE I Inferrino Relationships Between Indigenous Baja California Sur and Serl/Comáac Populations Through Cultural Traits | Macfarlan / Henrickson

the overall pattern was likely correct. Additionally, identified by historical explorers and Jesuit missionaries, he suggested that Kroeber's trait list be revised and as well as by cultural anthropologists, archaeologists, expanded to discern relationships between BCS groups and historians who have examined or translated the and the Seri/Comcáac. To our knowledge, no one has yet ethnohistoric documents. attempted that task. There are multiple ways to derive lists of cultural In this paper we introduce a method, content analysis, traits. Archaeologists derive them from the material which can render the anecdotal nature of historical record; cultural anthropologists generally use ethno documents into systematic cultural trait distribution graphies. lists. Cross-cultural anthropologists commonly Once constructed, cultural trait data allow hypotheses employ to distribution lists of cultural traits to test relational be tested about macro-level social interactions between hypotheses about the ecological, historical, or social groups (Ryan and Bernard 2000). We use reliability correlates of human behavior (e.g., Barry and Schlegel and hierarchical cluster analysis to determine cultural 1980). There are drawbacks to this type of research (also relationships within BCS populations, and between termed holocultural research). For example, cultural trait them and the Seri/Comcáac. Because there is a dearth lists generally are based on a few descriptive sentences of both theory and data to aid in predicting relationships about the presence or absence of a particular trait in a between cultural groups in southern BCS, our analysis particular culture at one given time. Additionally, coding is largely exploratory. Based on our analyses, it appears the presence or absence of a trait for a cultural group based that (1) the historic Guaycura and southern Cochimi on information from a limited number of consultants were culturally very similar; (2) the degree to which masks the all of the variability that often exists within a culture Guaycura and southern Cochimi shared cultural (Hewletttraits and Macfarlan 2010), particularly with regard to was much greater than any other grouping; (3) the age, Seri/ gender, and status. Given these limitations, however, Comcáac formed a distant but natural grouping with holocultural the research can still be useful for determining Guaycura and the southern Cochimi; and (4) the broadPericú patterns between cultures. were culturally distinct. Two reviewers (the authors) developed a list of cultural traits based on four classes of source material: (1) translated historical documents related to explorations of METHODS Baja California spanning the period of A.D. 1539-1721 This study utilizes classical content analysis (Cooke (Ryan 1992; and de Alarcón 1992; de Atondo y Antillón and Bernard 2000) to investigate relationships between Kino 1992; three de Cardona 1992; de la Ascension 1992; de la historical BCS cultural groups: the Pericú, Nava Guaycura, 1992; de Lucenilla 1992; de Ortega 1992; de Ulloa and the southern Cochimi, as well as the Seri/Comcáac 1992a, 1992b; Porter y Casanate 1992; Shevlocke 1992; of the Gulf of California and Sonora Mexico Vizcaíno (Fig. 1992a, 1). 1992b); (2) translated Jesuit missionary Content analysis uses messages (e.g., texts), ratheraccounts than of Baja California Sur groups spanning the behavior or artifacts, as the unit of study period (Neuendorf of A.D. 1683-1768 (Baegert 1952; Burrus 1984; del 2002). Although content analysis involves Barco a range 1981; ofNunis 1982); (3) Seri/Comáac ethnographic techniques, the basic premise is the same — accountsresearchers (Bowen 2000; Feiger and Moser 1991; Kroeber convert qualitative texts into quantitative 1931;data, McGeewhich 1898) and archaeological research in can then be used to test relational hypotheses Seri/Comáac (Ryan and territory (Bowen 1976); and (4) peer Bernard 2000). Content analysis requires (1)reviewed selecting academic research pertaining to historical BCS texts for analysis; (2) defining the variables (Aschmannto be coded; 1967; Heizer and Massey 1953; Kroeber (3) applying those codes systematically to a set1931; of Laylandertexts; 2007; Massey 1947,1949, 1961,1966; (4) testing the reliability of coders when more Mathes than 1992,2006). one As an initial exploration of a method is present; (5) creating a unit-of-analysis-by-trait involving matrix time-consuming research, we chose not to from the texts and codes; and 6) hypothesis include testing information from groups living north of the 28th using statistical methods (Bernard 2002). The parallel traits in thewe peninsula, since this approximates the examined are institutional structures and ethnic markers northern boundary of the southern Cochimi language

(Laylander 1997; Mixco 2006), and groups below this more accurate cultural knowledge about the groups they boundary had had definite, documented contacts with discuss. Europeans since A.D. 1539 (Mathes 1981). We chose to exclude the Cora and Monqui because insufficient data RESULTS existed for a separate trait analysis, and their group and linguistic affiliations were ambiguous. Due to human error (e.g., an incorrect reading of texts or data coding), it is important to evaluate inter-coder Assigning Traits to Groups agreement or reliability (Ryan and Bernard 2000). A list of 88 potential cultural traits was derived from Reliability is determined by whether or not a measuring initial readings of Baegert (1952), del Barco (1981), procedure yields the same results on multiple trials Burrus (1984), and Nunis (1982). These traits largely (Carmines and Zeller 1982), and it is evidence that a describe Guaycura and southern Cochimi culture, and coded theme has some external validity (i.e., is not a (to a lesser extent) that of the Pericú. Cultural traits figment of the researcher's imagination) (Ryan and pertaining to the Seri/Comcáac were easily derived, as Bernard 2000). Therefore, a reliability analysis was used trained ethnographers and linguists have researched to determine the degree of accuracy between raters. these groups since the pioneering work of McGee (1898). Conventions in reliability analysis are varied; however, Pericú cultural traits were the most difficult to code many authors agree that coefficients greater than 0.7 due to a lack of data; however, historical accounts from are sufficient for exploratory research to be followed by explorers spanning the period of A.D. 1537-1712 weresubsequent analyses (Landis and Kosh 1977). High inter vital for purposes of reconstruction. This process resulted rater agreement was achieved for the traits assigned to in our final data set being reduced to 51 cultural traits, the four cultural groups (Pericú: Cohen's K=0.95, N=51, grouped into five categories: (1) male headdress; (2)pcO.OOl; Guaycura: Cohen's K=0.8, N=51, p<0.001; female dress; (3) religious practices/marriage; (4) childsouthern Cochimi: Cohen's K=0.92, N=51, p< 0.001; carrying devices; and (5) technology (see Table 1). Seri/Comcáac: Cohen's K=0.92, N=51, p< 0.001). When We used a presence/absence dichotomy in assigning disagreements occurred regarding a trait's proper coding, cultural traits to groups. Because we did not want the to primary source material was reviewed and the bias our sample, we chose to use a trait only if sufficient appropriate scheme determined through consensus. Thus, information existed for all four groups. Our rationale thewas consensus-building process eventually resulted in that if an account identified a trait for one group but perfect no agreement between the raters for all of the traits information was recorded concerning the other groups, for all four cultures. marking this trait as absent could inflate the similarity of Reliability and hierarchical cluster analyses were the other groups when statistical analyses were run. Some employed to determine data structure. Reliability analysis traits were recorded as "not present" when an alternative determines a set of items' internal consistency when version was present and no author made a claim about measured with Cronbach's alpha (Vogt 2005). When the trait's absence (indicated by an askerisk in Table 1). items are cultural traits, reliability analysis determines the We used a two-tiered system to reconcile instances extent to which groups share a culture. High reliability where accounts/authors differed on the presence/absence coefficients (e.g., >0.7) indicate that groups share a of a trait. If a trait was suggested by one account common to culture. Low reliability coefficients indicate that be absent, but another recorded its occurrence, we groups are culturally distinct from one another. A low sided with the author who reported the presence. Our reliability coefficient was attained when all four cultural rationale was that it is easier to mistakenly identify a trait groups were examined simultaneously (Cronbach's as being absent than as being present. Secondly, we gave a = 0.40; N=51). A second set of reliability analyses more weight to accounts from authors who had spent a were run examining three cultures simultaneously; greater amount of time with a group than we did to those these resulted in moderate to extremely low reliability from authors who had never visited the peninsula or coefficients (Guaycura-southern Cochimi-Seri: visited only briefly. We felt that the former should have Cronbach's a = 0.47; N=51; Pericú-Guaycura-southern

This content downloaded from 142.103.160.110 on Fri, 11 Jan 2019 05:39:20 UTC All use subject to https://about.jstor.org/terms ARTICLE I Inferring Relationships Between Indigenous Baja California Sur and Serl/Comáac Populations Through Cultural Traits | Maofarlan / Henrickson

Table 1

CULTURAL TRAIT LIST (1=PRESENT; O=ABSENT)

Southern Category Cultural Trait Pericú Source Guaycura Source Source Seri Source Cochimf

Male Pearl Knotted del Barco 1 del Barco 1981:37 0 del Barco 1981:37 0 0 McGee 1898:101 Headdress in Hair 1981:38

Cooke 1992:295; Porter Ulloa 1992b:79; y Casante 1992:249; de Kroeber 1931:43; Baegert del Barco Kroeber 1931:43; Long Hair 1 1 0 1 Atondo and Kino 1992:167; 1982:139, photos; 1981:38 McGee 1898:photos del Barco 1981:37 Baegert 1952:88

Porter y Casante Hair Top 1992:249; de la del Barco McGee 1898:101; 1 0 del Barco 1981:37 0 0 Knotted Ascension 1992:167; 1981:38-39 Kroeber 1931:44 Mathes 2006:51

Feiger and Moser del Barco Crown 1 del Barco 1981:39 1 Ulloa 1992b:82 1 1 1991:146; McGee 1981:38 1898:101 disagrees

Net Worn del Barco 1981:37; Ulloa del Barco 0 Mathes 2006:52 1 0 0 Kroeber 1931:44 on Head 1992b:82 1981:38-39

del Barco

Mother-of Mathes 2006:51; 1981:38-39; ♦ 1 1 del Barco 1981:37 1 0 Pearl in Hair Shelvocke 1902:320 de Antondo and Kino 1992:33

del Barco 1981:43-44; del Barco - Female Dress Palm Skirt 1 del Barco 1981:40 0 0 0 Baegert 1952:62 1981:43 Cardona 1992:102 Mammal (island),216; Feiger del Barco Leather Use 1 Cooke 1992:294 1 Baegert 1952:62 1 1 and Moser1991:45; 1981:69 for Skirt Hardy 1977:298; McGee 1898:86,92

* * McGee 1898:92; Bird Hide Skirt 1 Cooke 1992:294 0 0 1 Kroeber 1931:22

Agave Fiber del Barco 1981:43-44; del Barco * 0 del Barco 1981:40 1 1 0 Skirt Baegert 1982:138 1981:42

Baegert 1952:62; Kroeber 1931:43; del del Barco Two Part Skirt 1 del Barco 1981:40 1 1 0 Kroeber 1931:43 Barco 1981:43 disagrees; 1981:42 Baegert 1982:138

de Salvatierra 1971:107; Reed Beads * del Barco 0 1 del Barco:1981:43; 1 0 Kroeber 1931:43 Skirts 1981:42 Baegert 1952:61

Leather Skirt ♦ Baegert 1952:62; 0 1 1 Kroeber1931:43 0 Kroeber 1931:43 on Back Only Baegert 1982:138

Ulloa 1992a, Baegert 1982:140-141, McGee 1898:139; Long Hair 1 del Barco 1981:41 0 0 1992b; 1 photos; Baegert 1952:88 Kroeber 1931:43 Mathes 2006:52

Bird Hide used Baegert 1952:62; del Barco McGee 1898:86; Feiger 1 Shelvocke 1992:323 0 0 1 in Clothing Baegert 1982:138 1981:41-42 and Moser 1991:149

Shelvocke 1992:323; Cape/shirt, del Barco McGee 1898:86,92; 1 de la Ascension 1992:168; 0 Baegert 1952: 61-62 0 0 everyday wear 1981:41-42 Kroeber1931 del Barco 1981:40

Table 1 (Continued)

CULTURAL TRAIT LIST (1=PRESENT; 0=ABSENT)

Southern Category Cultural Trait Perieli Source Guaycura Source Source Seri Source Cochimf

de Antondo and McGee 1898:279; Religious Kino 1992:275; del Barco 1981:49; del Barco Bowen 2000:21,471; Practices/ Polygynous 1 1 Baegert 1952:73; 1 1 Mathes 2006:58 1981:49 Kroeber 1931:8,24, 43 Marriage del Barco 1981:49; disagrees Mathes 2006:58

Shaman Aschmann Kroeber 1931:43; Human Hair 1 Kroeber1931:44 1 1 1967:114; 0 Kroeber 1931:43, 44 Baegert 1952:88 Cape Kroeber1931:44 Shaman ♦ Kroeber 1931:43; Aschmann Sucking 0 1 1 0 Kroeber1931:43 Baegert 1952:78 1967:113 Implement

de la Ascensión 1992:188; Face/Body Ulloa 1992b;83; Mathes 1974:103, 1 Mathes 2006:51; Porter y 1 1 1 Kroeber 1931:40 Painting Baegert 1952:62 105 Casante 1992:249

DiPeso and Matson

Face paint Mathes 2006:51; * 1965:51-52; Feiger and 1 0 Baegert 1952:62 0 1 is non-Ritual Shelvocke 1992:323 Moser 1991:144,152; Kroeber 1931:27

Shamans * Aschmann 0 1 Mathes 2006:65 1 0 Kroeber 1931:44 Tablas 1967:115-116

Shamans Kroeber 1931:42; 1 Kroeber1931:42 1 1 Kroeber1931:42 1 Kroeber 1931:42, 43 Use Caves Mathes 2006:85

Hair Cut in Kroeber 1931:43; 1 Ortega 1992:243 1 1 Mathes 1974:105 1 Kroeber 1931:43 Mourning Baegert 1952:88

Kroeber 1931:44; Child Carrying Baegert 1982:142; Feiger and Moser 1991: Net/Bag 0 Kroeber 1931:44 1 1 del Barco 1 Device del Barco 1981:71-72 139 1981:71-72

Kroeber 1931:41; Tray/Turtle * 1 del Barco 1981:72 1 Baegert 19 5 2:63,74; 0 0 Kroeber1931:41, 42 Shell Baegert 1982:142

* * » Kroeber 1931:40; Stick Cradle 0 0 0 1 McGee 1898:226

Attached to del Barco 1 del Barco 1981:72 1 Baegert 1982:142 1 0 Kroeber 1931:44 Forehead 1981:71-72

Attached to • * Mathes 2008:55, Feiger and Moser 0 0 1 1 Pole/Yoke 59 1991:190

Fire Hardened * Technology 1 Laylander 2007:15-17 1 Laylander 2007:15-17 0 1 McGee 1898:190 Poles

Lances/ Ulloa 1992b:79: * 1 Ortega 1992:230; 1 1 Laylander 0 Javelins Laylander 2007:15-17 Laylander 2007:15-17 2007:15-17

Ortega 1992:230; * * Harpoons 1 0 0 1 Kroeber 1931:43 Laylander 2007:15-17

Feiger and Moser Laylander 2007:15-17; * Fishing Spear 1 Laylander 2007:15-17 1 0 1 1991:128; Kroeber 1931: Baegert 1952:176 19,40; McGee 1898:193

Kroeber 1931:44; Aschmann 1967:73; Fishing Nets/ Massey 1966:54; Kroeber 1931:19, 44; 1 Ortega 1992:230 1 Mathes 2006:55; 1 0 Traps Laylander 2000; McGee 1898:194 Laylander 2000:127 128

♦ Atlatl 1 Massey 1961 0 Massey 1961 0 Massey 1961 0

This content downloaded from 142.103.160.110 on Fri, 11 Jan 2019 05:39:20 UTC All use subject to https://about.jstor.org/terms ARTICLE ¡ Inferring Relationships Between Indigenous Baja California Sur and Seri/Comáac Populations Through Cultural Traits | Macfarlan / Henrickson

Table 1 (Continued)

CULTURAL TRAIT LIST (1=PRESENT; 0=ABSENT)

Southern Category Cultural Trait Perieli Source Guaycura Source Source Seri Source Cachimi

Cooke 1992:295; Porter y Technology de Salvatierra 1971:120; Aschmann Feiger and Moser Bow and Arrow 1 Casante 1992:249; Ortega 1 1 1 (Continued) Baegert 1952:63 1967:66 1991:126 1992:230

Kroeber 1931:43; Arrow foreshaft Baegert 1952:64; 1 Cooke 1992:296 1 1 Mathes 2006:56 1 Kroeber 1931:40 is cane reed Ulloa 1992b:78; Mathes 2006:55

Kroeber1931:43; Baegert 1952:64; Kroeber1931:40; Long bows 1 Cooke 1992:295 1 1 Mathes 2006:50 1 Baegert 1982:139; McGee 1898:200 Ulloa 1992b:78

Feiger and Moser Non-Ritual Cooke 1992:294; del Barco 1981:46; 1991:118-120; Laylander Roofed 1 Cardona 1992:224; 0 Baegert 1952:59; 1 1 Kroeber1931:43; 2000:154 Housing del Barco 1981:45 Baegert 1982:142 Bowen 1976:45; McGee 1898:221-224

Feiger and Moser Unroofed Rock * Aschmann 0 1 Kroeber 1931:42 1 1 1991:119-120; Windbreak 1967:108-110 Kroeber 1931:40

Feiger and Moser 1991:180; Basketry 1 Ortega 1992:230 0 Massey 1966:56 1 Massey 1966:54 1 Kroeber 1931:40; McGee 1898:208

Aschmann Coiled * • Kroeber1931:40; 0 0 1 1967:62; 1 Basketry McGee 1898:208 Massey 1966:54

Massey 1966: * Massey 1966; Feiger and Moser 1991:9; Pottery 0 0 0 Aschmann 1967: 1 Aschmann 1967:38 Kroeber 1931:40 38

* • * Agriculture 0 0 0 0 Kroeber1931:47

Feiger and Moser Female Head Kroeber1931:40; 1991:139; 0 Kroeber 1931:40 0 0 Kroeber 1931:40 1 Carrying Baegert 1952:88 McGee 1898:149; Kroeber 1931:16

Agave Fiber Kroeber1931:41; Aschmann 1967: 1 Mathes 2006:54 1 1 1 Kroeber 1931:40 Nets Baegert 1952:68 73

Second Kroeber 1931:40; McGee Harvest of 1 del Barco 1981:77 1 Baegert 1982:144 1 del Barco 1981:77 1 1898:209 Pitahaya

Shell for Shelvocke 1992; del Barco 1 1 del Barco 1981:37 1 1 McGee 1898:173 Adornment Mathes 2006:52 1981:38-39

Feiger and Moser Baegert 1952:82; Heizer and 1Φ)1·1Ί1· Reed Balsa 1 Heizer and Massey 1953 1 Kroeber 1931:42; 1 Massey 1953; 1 Kroeber1931:40; Boat Massey 1947 Massey 1947 McGee 1898:216-219; Ulloa 1992b:78

Table 1 (Continued)

CULTURAL TRAIT LIST (1=PRESENT; 0=ABSENT)

Southern Category Cultural Trait Perieli Source Guaycura Source Source Seri Source Gochimi

Cooke 1992:295; Technology Heizer and Massey Heizer and Heizer and Massey Log Boats 1 Ortega 1992:229; 0 0 0 (Continued) 1953:290 Massey 1953:290 1953:290 Massey 1947

Cooke 1992:295; Heizer and Double Bladed Heizer and Massey 1953; Kroeber 1931:40; 1 Heizer and Massey 1953; 1 1 Massey 1953; 1 Paddle Massey 1947 McGee 1898:219 Massey 1947 Massey 1947

Cochimí: Cronbach's α=0.47; N=51; Pericú-Guaycura natural classification of a set of objects is unknown and Seri: Cronbach's α = 0.15; N=51; Pericú-southern taxonomic order is desired (Aldenderfer and Blashfield Cochimí-Seri: Cronbach's α=0.17; N=51). A final set of 1984). Hierarchical cluster analysis is one clustering reliability analyses examined pairs only. High internal technique that places single entities into increasingly reliability was reached for the Guaycura and southern homogeneous groupings using an iterative process. Cochimí (Cronbach's a= 0.77; N=51); however, extremely Although standards vary, many scholars agree that low or negative reliability coefficients were derived for all hierarchical cluster analysis is a preferred clustering other pair-wise groupings (Table 2). Negative reliability method for small sample sizes (e.g., <250 cases), with a coefficients are indicative of small sample sizes or the minimum requirement of no less than 2k cases (k=number evaluation of multiple constructs (Krus and Helmstadter of variables) (Dolnicar 2002). Hierarchical clustering 1993)—i.e., different cultures. Although the sample is requires a similarity metric to assess distances between moderately small, it appears that multiple cultures were groups and a link-function to hierarchically organize examined simultaneously. Given the high cultural trait them. It is vital to have a justification for selecting one agreement between southern Cochimí and Guaycura, similarity metric and one link-function over others, as the additional constructs being evaluated are the Pericú output is determined by these choices (Aldenderfer of the Cape Region and the Seri/Comcáac cultures ofand Blashfield 1984). We chose a Phi 4-point correlation mainland Mexico. similarity metric and a within-groups link function. The Phi 4-point correlation procedure was selected over other binary data similarity measurements because of its ease Table 2 of interpretability (it is equal to the Pearson product PAIR-WISE RELIABILITY COEFFICIENTS moment correlation coefficient for binary data), and it BASED ON CULTURAL TRAITS gives equal weight to the joint presence and absence of Southern traits to calculate similarity. Because we selected traits Perieli Guaycura Cachimi' Seri where the joint absence of a trait was as meaningful Pericú - 0.13 -0.09 0.24 as its presence, this metric was more appropriate than Guaycura - - 0,7/ -0.08 those that exclude joint absences from computation Cochimí - - 0.17 (Aldenderfer and Blashfield 1984). The within-groups link function was selected because it was designed for the specific purpose of determining homogeneity within Due to the moderate clusters reliability by an examination estimates of for both inter- and intra the Guaycura-southern cluster pairsCochimi-Seri (Garson 2009). and GuaycuraThis resulted in two clusters: southern Cochimí-Pericú (1) the geographically groupings, adjacentwe sought southern to Cochimi and determine whether Guaycura deeper of structures the south-central existed peninsula within form a distant the data. Researchers, yet single including group anthropologists with the Seri/Comcáac (e.g., of mainland Maxwell et al. 2002), Mexico; employ while cluster (2) the analysisPericú of when the southern the peninsular

This content downloaded from 142.103.160.110 on Fri, 11 Jan 2019 05:39:20 UTC All use subject to https://about.jstor.org/terms ARTICLE I Inferring Relationships Between Indigenous Baja California Sor and Seri/Comáac Populations Through Cultural Traits | Macfarlan / Henriokson

tip were isolated. Identical results were obtained using cluster stability can become problematic. If new traits other similarity metrics (i.e., Lambda, Anderberg's D, are identified and added to our list, it is possible the and Yule's Q). groupings we identified will no longer be meaningful. Thus interpretations must proceed with caution. Second, one may question the utility of the DISCUSSION particular cultural traits identified. Many were based on The present study was performed to systematize observations cultural by people with no ethnographic training. trait information from ethnohistoric documents fromAdditionally, BCS sparse data exist for Pericú institutional and to formally analyze cultural relationships elements between (e.g., descent rules, marriage patterns), which the Pericú, Guaycura, southern Cochimi, and are theexactly Seri/ those traits preferred for analysis by cross Comcáac populations through inferential statistics. cultural The anthropologists (e.g., Barry and Schlegel 1980). former was accomplished through content However, analysis, many of the traits we identified appear to be while the latter was carried out through reliability ethnic markers and (e.g., style of hair and dress). These can hierarchical cluster analysis. Based on the traits be identified, desirable data points because they are maximally it appears that (1) the historic Guaycura and arbitrary; southern thus when groups share these traits it represents Cochimí shared a similar culture; and (2) the Seri/Comcáac some shared cultural schema for behavior (Brown 2008; formed a more natural grouping with the Strauss Guaycura and Quinn 1997). and southern Cochimi than did the Pericú. Although Finally, the characterization of each cultural group largely exploratory in nature, these analyses support is a composite some picture formed by accounts spanning assumptions about how certain historical BCStime, groups geographic locations, and levels of cultural contact were culturally related to one another, while withweakening Europeans. Thus the trait list derived for each others. First, we reinforce Massey's (1947) assumption group is thatunique in some way. Traits identified for the the Cochimi and the Pericú were culturally distinct Seri/Comcáac from are derived from ethnographic source one another. Secondly, we reinforce the assumption materials that spanning three locations (Isla Tiburón, the Pericú were culturally distinct from nearly Isla allSan other Esteban, and Coastal Sonora) from the late groups. We reject Massey's (1947) assumption nineteenth that the and early twentieth centuries. Pericú and Guaycura were as culturally similar to the Pericú some as Guaycuran to the traits were retrieved from historical Cochimi. Finally, we reject Kroeber's (1931) interpretation accounts spanning the early sixteenth to eighteenth that the Guaycura had a greater cultural similarity centuries. to the Jesuit missionary accounts provided many Seri/Comcáac than the Pericú or Cochimi. Our results are traits for the Guaycura and southern Cochimi, but discussed below with reference to analytical limitations proved more difficult to use for identifying Pericú traits and historical processes. (e.g., Napoli 1970). In general, these accounts are not as complete as those that originate north of the cape Analytic Limitations region. Consequently, one must ask whether the close Several data-level and analysis-related features require associations found between the southern Cochimi and attention before giving full consideration to the Guaycura, and the more distant associations of the historical and cultural mechanisms affecting inter-group Pericú and the Seri/Comcáac, are simply a byproduct of relationships. First, we sought to create a cultural trait list the temporal periods when the traits were recorded. One that was sufficiently large to test relationships between might reasonably ask, if Jesuit missionaries had recorded groups; however, the list is not exhaustive. Documents as much cultural information for the Seri/Comcáac that have not been translated were not analyzed, some during the seventeenth and eighteenth centuries A.D. historical documents were inaccessible because of their as they did in the peninsula, would the Seri/Comcáac location (e.g., in remote repositories), and archaeological appear to be more similar culturally to the Guaycura and investigations are ongoing. Therefore, it is likely that southern Cochimi? Despite these limitations, this is the additional traits will be identified in the future. When most complete picture derivable for these cultural groups cluster analyses are run on small samples (like ours), at this time and we consider it appropriate to directly

assess the implications of our findings through several dissimilar to the Guaycura and southern Cochimi. cultural processes, discussed below. The mixed marine/terrestrial foraging economy of the southern Cochimi and Guaycura was distinct from the Cultural Processes largely aquatic foraging economies and coastal habitat of This is the first study that we are aware of that has the Seri/Comcáac (Bowen 2000) and Pericú. However, formally evaluated the direction and magnitude of convergent processes are insufficient, as they explain only cultural similarities and differences within historical BCS ecologically salient traits. cultures, and between these groups and Seri/Comcáac The third process is plausible, given the pattern of populations, through inferential statistics. The Guaycura European and indigenous actions on the Baja California and the southern Cochimi appear to have been culturally peninsula during the historic period. Explorers prior to homogeneous, if one uses a composite of cultural traits the mission period were enlisted by the Pericú to attack derived from historic accounts spanning the sixteenth the Guaycura on several occasions (de Ortega 1990; through eighteenth centuries. They had more in common Massey 1966; ). Indeed, many accounts explicitly state with one another than they did with the Pericú or the that the Pericú and Guaycura were at war, possibly due Seri/Comcáac. Although hierarchical cluster analysis to an incursion of the Guaycura into the Cape Region of suggests a natural grouping with the Seri/Comcáac, BCS in order to gain access to its preferential resource reliability coefficients suggest the Seri/Comcáac and base (Massey 1966) around A.D. 1670 (Mathes 1975) or the Pericú were virtually equidistant culturally from the earlier; however, Laylander (1997:16) disagrees with this southern Cochimi and Guaycura. This pattern could be interpretation. The low reliability coefficients derived for the result of at least three processes: (1) the Guaycura the Guaycura and Pericú indicate considerable cultural and Cochimi descended from a similar linguistic stock differentiation despite their geographic proximity. and shared an historical trajectory, so that their cultures Because ethnic markers signal in-group membership, were more similar to each other than either was to any warfare may have caused these groups to diversify along other group; (2) their descent was ambiguous; however, these dimensions. Guaycura peoples may have sought their shared ecology and forager lifestyle constrained cultural contact with southern Cochimian groups, in light their cultural repertoires to develop in parallel, without of the Pericú's enlistment of European person- and fire cultural contact; or (3) their descent was ambiguous, power to resist intrusion. but recent cultural diffusion had caused them to appear Interestingly, the Guaycura and southern Cochimi culturally homogenous. share all religious and female dress traits. This suggests The first potential process seems unlikely, given females may have moved exogamously between groups, recent linguistic evidence suggesting that the Guaycura thus possibly sharing religious ideas and female attire. and Cochimi languages are not genetically related Three pieces of information provide tangential evidence (Mixco 2006). Even if these groups shared a common regarding this question: (1) missionary reports suggest linguistic history, as suggested by Gursky (1966) and that only Cochimi men could make basketry (del Barco Swadesh (1967), it would likely be in the remote past, 1981); (2) there is no indication either archaeologically as an extension of inclusion in a larger Hokan language (Massey 1966) or historically (Baegert 1952) that the stock. That deep linguistic ancestry is unlikely to have Guaycura made basketry; and (3) female shamans produced the cultural similarities between the Guaycura were present during the historic period among the and southern Cochimi that existed during the historic Guaycura and possibly the southern Cochimi (Baegert period. 1952; Massey 1966). Aschmann (1967) suggests the If the Guaycura and Cochimi did not share a Cochimi were patrilocal. If males remain amongst their common linguistic heritage, it is possible that the shared natal kin throughout life, male traits are less likely to desert environment and foraging economy resulted in be shared between groups. As a consequence, male their cultures becoming similar through convergent traits, such as Cochimian basketry construction, would mechanisms. This process has the additional benefit not have been shared with the Guaycura. Additionally, of explaining why the Pericú and Seri/Comcáac are females are known to have performed religious functions

among the Guaycura (Baegert 1952). If females moved cultural features via mechanisms of innovation and/or exogamously between groups, one would expect the drift (Neiman 1995), while the Guaycura more recently southern Cochimi and Guaycura to share religious came into contact with the southern Cochimi. elements and ethnic markers related to femininity. Questions regarding the relationships between the On face value this proposition is appealing; however, Pericú and other Native American groups have been the it will require archaeological data for confirmation. It subject of academic discourse since the late eighteenth seems plausible that convergent cultural evolutionary century (Massey 1947). The debate centers on whether the and diffusionary processes could have been working in Pericú represent a remnant or separate population of early tandem to produce these patterns. migrants into the New World (González-Jose et al. 2003), Hierarchical cluster analysis revealed that the one that remained isolated into historic times (Massey Guaycura, southern Cochimi, and Seri/Comcáac formed 1966) via a culturally marginalizing "peninsular effect." Our a more natural grouping with one other than any other data suggest that the Pericú were culturally distinct from hierarchical grouping with the Pericú. This suggests at the Guaycura and southern Cochimi during the period least two processes might have been at work: (1) the of historic contact. However, this does not address the Guaycura, southern Cochimi, and Seri/Comcáac shared issue of the antiquity of the Pericú, as cultural mechanisms a similar linguistic history, with the Seri/Comcáac being can cause groups to diverge quickly, especially when very distantly related, while the Pericú were relatively population sizes are small (Neiman 1995). Contemporary distinct culturally; or (2) the southern Cochimi and Seri/ academic use of the term marginality is relative, and it is Comcáac shared a distant history, and the Guaycura often employed without contextual reference to a core were recently engaged in sustained contact with the (cultural, political, economic, or geographic) (Turner and southern Cochimi, while the Pericú were culturally Young 2007). Our data do not address Pericú origins; distinct. The first process is unlikely, given the limited however, if marginality is defined as involving distinct linguistic evidence. Although evidence suggests the boundaries between cultural groups, it is likely that a Cochimi, Seri/Comcáac, and Yuman groups shared cultural boundary, in conjunction with a phytogeographic ancient linguistic features (Kroeber 1931; Mixco 1978, boundary, existed between the Pericú and the Guaycura 2006), it is unlikely the Guaycura were related (Mixco during the historic period that hampered contact, while 2006). Even if Guaycura were related to Cochimi and such a distinct cultural boundary did not exist between Seri/Comcáac through a shared Hokan language family the Guaycura and the southern Cochimi. affiliation, it would not explain the negative reliability coefficient between the Guaycura and Seri/Comcáac, CONCLUSIONS unless selection pressures shaped cultural traditions in radically different, yet locally relevant ways. This paper builds on a program of research that was The second process appears more likely. Some started by Kroeber (1931) and expanded upon by Massey linguistic evidence supports the idea that Cochimi and (1947), Laylander (1997, 2000), and Mathes (2006), Seri/Comcáac were distantly related (Kroeber 1931; a program that involves deriving cultural traits from Mixco 2006). Thus these groups might have shared historic documents in order to infer group relationships, an ancient history. Indeed, some interpretations of both between cultures in historical BCS and with cultural Seri/Comcáac oral history place their origins in central groups outside the region. This paper's contribution Baja California (Bowen 1976; Moser and White 1968). lies in the fact that it has tested assumptions about Additionally, Seri/Comcáac oral traditions (Bowen cultural relationships between groups through inferential 2000:23-25) suggest contact with coastal, central Baja statistics rather than through intuition alone. Cultural California peoples. Although our analysis suggests the traits comprise an important set of evidence that can Guaycura and Seri/Comcáac populations were culturally independently reinforce linguistic research in the dissimilar, both shared traits with the southern Cochimi. investigation of inter-group contacts. Archaeologists It seems possible that the Seri/Comcáac shared a distant and historians will play a vital role in creating larger, history with the southern Cochimi, but developed unique more meaningful cultural trait distribution lists for BCS

indigenous populations. Articulating these datasets Brown, M. J. 2008 When Culture Does Not Affect Behavior: The with information from cultural groups in the northern Structural Basis of Ethnic Identity. In Explaining Culture half of the peninsula, southern Alta California, and Scientifically, M. J. Brown, ed., pp. 162-183. Seattle: mainland Mexico will be fundamental in reconstructing University of Washington Press. the prehistory of these regions. Constructing cultural trait Burrus, E. J. distribution lists through written and material records, 1984 Jesuit Relations, Baja California, 1716-1772. [Baja in conjunction with linguistic data, allows one to test California Travel Series 47.] Los Angeles: Dawson's Book Shop. hypotheses about cultural or ecological marginality and core/periphery relationships, rather than simply assuming Campbell, L. their applicability to the Baja California peninsula. 1997 American Indian Languages: The Historical Linguistics of Native America. New York: Oxford University Press.

Carmines, E. G., and R. A. Zeller ACKNOWLEDGEMENTS 1982 Reliability and Validity Assessment. Beverly Hills: The authors thank Don Laylander and one anonymous Sage reviewer Publications. for their thoughtful comments on the manuscript. Additional Clavijero, F. J. thanks to Amy Gusick, Matthew Des Lauriers, and Loren 1789 Storia della California. Venice: Appresso Modeso Davis for organizing the SAA session that resulted in Fenzo. an earlier version of this manuscript, and to Lynn Gamble for inviting us to participate in this special Baja California edition. Cook, S. F. 1937 The Extent and Significance of Disease among the Indians of Baja California from 1697-1773. Berkeley: REFERENCES University of California Press.

Aldenderfer, M. S., and R. Κ. Blashfield Cooke, E. 1984 Cluster Analysis. Quantitative Applications 1992 [1712] in theCabo San Lucas: November-December 1709 Social Sciences 44. London: Sage Publications. (1712). In Ethnology of the Baja California Indians, M. Mathes, ed., pp. 289-305. [Spanish Borderland Aschmann, H. Sourcebooks 5.] New York: Garland Publishing. 1967 The Central Desert of Baja California: Demography and Ecology. Ibero-Americana 42. Berkeley: University Dalton, R. of California Press. 2005 Skeleton Keys. Nature 43:454-456. 1986 Learning about Baja California Indians. Journal of de Alarcón, F. California and Great Basin Anthropology 8:238-245. 1992[1540] Relation of Fernando de Alarcón: May-September Baegert, J. J. 1540([1600] 1904).Translated by R. Hakluyt. In Ethnology 1952 [ 1772] Observations in Lower California. Translated of the Baja California Indians, M. Mathes, ed., pp. 95-134. from German by M. M. Brandenburg and C. L. Baumann. [Spanish Borderland Sourcebooks 5.] New York: Garland Berkeley: University of California Press. Publishing. 1982 The Letters of Jacob Baegert, 1749-1761, Jesuit de Atondo y Antillón, I. de, and E. F. Kino Missionary in Baja California, Doyce B. Nunis, Jr., ed. Los 1992 [1683] Accounts of Isidro de Atondo y Antillón and Angeles: Dawson's Book Shop. Eusebio Francisco Kino, S. J.: 1683 (1990). Translated Barry III, H., and A. Schlegel by W. M. Mathes. Ethnology of the Baja California 1980 Cross-Cultural Samples and Codes. Pittsburgh: Indians, M. Mathes, ed., pp. 273-277. [Spanish Borderland University of Pittsburgh Press. Sourcebooks 5.] New York: Garland Publishing.

Bernard, H. R. de Cardona, Ν. 2002 Research Methods in Anthropology: Qualitative and 1992 [1615] Description of Nicolás de Cardona. Translated Quantitative Approaches. 3rd Ed. Walnut Creek: Altamira by W. M. Mathes. In Ethnology of the Baja California Press. Indians, M. Mathes, ed., pp. 209-220. [Spanish Borderland Bowen,T. Sourcebooks 5.] New York: Garland Publishing. 1976 Seri Prehistory: The Archaeology of the Central de la Ascension, A. Coast of Sonora, Mexico. Anthropological Papers of the 1992 [1602] Father Antonio de la Ascension's Account of University of Arizona 27.Tuscon: University of Arizona. the Voyage of Sebastián Vizcaíno: 1602 (1929). Translated 2000 Unknown Island: Seri Indians, Europeans, and San by H. R. Wagner. In Ethnology of the Baja California Esteban Island in the Gulf of California. Albuquerque: Indians, M. Mathes, ed., pp. 164-208. [Spanish Borderland University of New Mexico Press. Sourcebooks 5.] New York: Garland Publishing.

de la Nava, D. Foster, J.W. 1992 [1632] Opinion Given by Licentiate Diego de la Nava: 1984 A Late Period Seri Site from Bahía de los Angeles, April-June 1632 (1975). Translated by W. M. Mathes. In Baja California. Pacific Coast Archaeological Society Ethnology of the Baja California Indians, M. Mathes. ed., Quarterly 20(1): 61-68. pp. 221-228. [Spanish Borderland Sourcebooks 5.] New Garson, G. D. York: Garland Publishing. 2009 Cluster Analysis. Electronic document, http://faculty. del Barco, M. chass.ncsu.edu/garson/PA765/cluster.htm, accessed July 1981 Ethnology and Linguistics of Baja California. 17,2009. Translated by F. Tiscareno. [Baja California Travel Series González-José, R., A. González-Martín, M. Hernández, 44.] Los Angeles: Dawson's Book Shop. H. M. Pucciarelli, M. Sardi, A. Rosales, and S. Van der Molen. 2003 Craniometrie Evidence for Palaeoamerican Survival de Lucenilla, F. 1992 [1668] Relation of Francisco de Lucenilla: May-July in Baja California. Nature 425:62-65. 1668 (1966). Translated by W. M. Mathes. In Ethnology of Gursky, K. H. the Baja California Indians, M. Mathes, ed., pp. 254-272. 1966 On the Historical Position of the Waikuri. International [Spanish Borderland Sourcebooks 5.] New York: Garland Journal of American Linguistics 32(l):41-45. Publishing. Hardy, Lieutenant R. W. H. de Ortega, F. 1977 (1829) Travels in the Interior of Mexico in 1825,1826, 1992 [1634] Relations of Francisco de Ortega: 1633-1634 1827, & 1828. Facimile reprint of the 1829 edition. Glorieta, (1990). Translated by W. M. Mathes. In Ethnology of the New Mexico:The Rio Grande Press. Baja California Indians, M. Mathes, ed., pp. 229-244. Heizer, R. F., and W. C. Massey [Spanish Borderland Sourcebooks 5.] New York: Garland 1953 Aboriginal Navigation off the Coasts of Upper and Publishing. Baja California. Bureau of American Ethnology Bulletins de Ulloa, F. 151:285-311. Washington, D.C. 1992a [1600] Relation of Francisco de Ulloa: 1539([1600]1904). Hewlett, B., and S. J. MacFarlan Translated by R. Hakluyt. In Ethnology of the Baja 2010 Fathers' Roles in Hunter-Gatherer and Other California Indians, M. Mathes, ed., pp. 2-74. [Spanish Small-Scale Cultures. In The Role of the Father in Child Borderland Sourcebooks 5.] New York: Garland Development, 5th Edition, M. E. Lamb ed., pp. 413-434. Publishing. Hoboken: John Wiley & Sons. 1992b [1600] Voyage of Francisco de Ulloa: July 1539-May Horning, A. 1540 (1929). Translated by H. R. Wagner. In Ethnology 2007 Materiality and Mutable Landscapes: Rethinking of the Baja California Indians, M. Mathes ed., pp. 75-94. Seasonality and Marginality in Rural Ireland. Inter [Spanish Borderland Sourcebooks 5.] New York: Garland national Journal of Historical Archaeology 11:358-378. Publishing. Hugo, H., and E. Exequiel de Salvatierra, J. M. 2007 Endemic Regions of the Vascular Flora of the Peninsula 1971 Selected Letters about Lower California. Translated of Baja California, Mexico. Journal of Vegetative Science and annotated by Ernest J. Burrus, S.J. Los Angeles: 18:327-336. Dawson's Book Shop. Hyland, J. R. DiPeso, C. G, and D. S. Matson 1997 Image, Land, and Lineage: Hunter-Gatherer Arch 1965 The Seri Indian in 1692 as Described by Adamo Gilg, aeology in Central Baja California, Mexico. Ph.D. S.J. Arizona and the Weit 7(1)33-56. dissertation, University of California, Berkeley.

Dolnicar, S. Jackson, R. H. 2002 A Review of Unquestioned Standards in Using 1983 Patterns of Demographic Change in the Missions of Cluster Analysis for Data-Driven Market Segmentation. Southern Baja California. Journal of California and Great Paper read at the annual meeting of the Australian and Basin Anthropology 5(1—2):131—139. New Zealand Marketing Academy, Deakin University, 1984 Demographic Patterns in the Missions of Central Melbourne. Baja California. Journal of California and Great Basin Anthropology 6(1):91-112. Feiger, R. S., and M. B. Moser 1991 People of the Desert and Sea: Ethnobotany of the Seri Jordan, P., and S. Shennan Indians. Tuscon: University of Arizona Press. 2003 Cultural Transmission, Language, and Basketry Traditions amongst the California Indians. Journal of Fernandez de Miranda, M.Teresa Anthropological Archaeology 22(l):42-74. 1967 Inventory of Classificatory Materials. In Handbook of Middle American Indians 5, R. Wauchope, ed., pp. 63-78. Kroeber,A. L. Austin: University of Texas Press. 1931 The Seri. Southwest Museum Papers 6. Los Angeles.

Krus, D. J., and G. C. Helmstadter McGee,W. J. 1993 The Problem of Negative Reliabilities. Educational 1898 The Seri Indians. Seventeenth Annual Report of the and Psychological Measurement 53:643-650. Bureau of American Ethnology, Part I.Washington, D.C.: Smithsonian Institution. Landis, J. R., and G. G. Koch 1977 The Measurement of Observer Agreement for McElreath, R., R. Boyd, and J. Richerson Categorical Data. Biometrics 33:159-174. 2003 Shared Norms and the Evolution of Ethnic Markers. Current Anthropology 44(1 ):122—129. Laylander, D. 1997 The Linguistic Prehistory of Baja California. In Mixco, M. J. Contributions to the Linguistic Prehistory of Central and 1978 Cochimi and Proto-Yuman: Lexical and Syntactic Baja California, G. S. Breschini and T. Haversat, eds., pp. Evidence for a New Language Family in Lower California. 1-94. Salinas: Coyote Press. University of Utah Anthropological Papers 101. Salt Lake City. 2000 Early Ethnography of the Californias, 1533 -1825. Salinas: Coyote Press. 2006 The Indigenous Languages. In Prehistory of Baja California: Advances in the Archaeology of the Forgotten 2007 Large Projectiles and the Cultural Distinction of Peninsula, D. Laylander and J. D. Moore, eds., pp. 24-41. Southern Baja California: A Reexamination. Pacific Gainesville: University of Florida Press. Coast Archaeological Society Quarterly 39(2&3):11-21. Moser, E., and R. S. White, Jr. Massey,W. C. 1968 Seri Clay Figurines. The Kiva 33(3):133—154. 1947 Brief Report on Archaeological Investigations in Baja California. Southwestern Journal of Anthropology Napoli, I. M. 3(4):344-359. 1970 [1721] The Cora Indians of Baja California. Translated by J. R. Moriarty III and B. F. Smith. Los Angeles: 1949 Tribes and Languages of Baja California. South Dawson's Book Shop. western Journal of Anthropology 5(3):272-307. 1955 Culture History in the Cape Region of Baja California, Neiman, F. D. Mexico. Ph.D. dissertation, University of California, 1995 Stylistic Variation in Evolutionary Perspective: Berkeley. Inferences from Decorative Diversity and Inter 1961 The Survival of the Dart-Thrower on the Peninsula of assemblage Distance in Illinois Woodland Ceramic Assemblages. American Antiquity 60(l):7-36. Baja California. Southwestern Journal of Anthropology 17(1):81—93. Neuendorf, K. A. 1966 Archaeology and Ethnohistory of Lower California. 2002 The Content Analysis Guidebook. Thousand Oaks: In Ethnology of the Baja California Indians, M. Mathes, Sage Publications. ed., pp. 334-361. [Spanish Borderland Sourcebooks 5.] Nunis, D. B. New York: Garland Publishing. 1982 The Letters of Jacob Baegert 1749-1761: Jesuit Mathes, M. Missionary in Baja California. Translated by E. Schulz 1974 Californiana III: Documentos para la Historia de la Bischof. Los Angeles: Dawson's Book Shop. Transformación Colonizadora de California, 1679-1686.3 O'Neil, S. vols. [Colección Chimalistac 36-38.] Madrid: José Porrúa 2008 Cultural Contact and Linguistic Relativity Among the Turanzas. Indians of Northwest California. Norman: University of 1975 Some New Observations Relative to the Indigenous Oklahoma Press. Inhabitants of La Paz, Baja California, Sur. Journal of Porter y Casanate, P. California Anthropology 2(2):180-182. 1992 [1644] A Spanish Voyage to California in 1644: The 1981 Problems of Ethnohistorical Research in Baja Report of Pedro Porter y Casanate (1976). Translated California. Journal of California and Great Basin by W. M. Mathes. In Ethnology of the Baja California Anthropology 3(l):44-48. Indians, M. Mathes, ed., pp. 245-272. [Spanish Borderland 1992 A Case of Idolatry Among the Cochimi (1974). In Sourcebooks 5.] New York: Garland Publishing. Ethnology of the Baja California Indians, M. Mathes, ed., Ritter, E.W. pp. 278-288. [Spanish Borderland Sourcebooks 5.] New 2006 South-Central Baja Caliornia. In Prehistory of Baja York: Garland Publishing. California: Advances in the Archaeology of the Forgotten 2006 Ethnohistoric Evidence. In Prehistory of Baja Peninsula, D. Laylander and J. D. Moore, eds., pp. 24-41. California: Advances in the Archaeology of the Forgotten Gainesville: University of Florida Press. Peninsula, D. Laylander and J. D. Moore, eds., pp. 42-81. Ryan, G. W., and H. R. Bernard Gainesville: University of Florida Press. 2000 Data Management and Analysis Methods. In Maxwell, Β. Α., E L. Pryor, and C. Smith Handbook of Qualitative Research, N. K. Denzin and 2002 Cluster Analysis in Cross-Cultural Research. World Y. S. Lincoln, eds., pp. 769-802. Thousand Oaks: Sage Cultures 13(l):22-38. Publications.

This content downloaded from 142.103.160.110 on Fri, 11 Jan 2019 05:39:20 UTC All use subject to https://about.jstor.org/terms ARTICLE 1 Inferring Relationships Between Indigenous Baja California Sur and Seri/Comáac Populations Through Cultural Traits | Macfarlan / Henrickson

Rogers, M. Tuohy, D. 1945 An Outline of Yuman Prehistory. Southwestern 1978 Cultural History in the Comondu Region, Baja Cali Journal of Anthropology 1:167-198. fornia, Mexico: With An Appendix on Metate Cave String Strauss, C, and N. Quinn and Twine Analysis by Carolyn M. Osborne. Master's 1997 A Cognitive Theory of Cultural Meaning. Cambridge: thesis, University of Nevada, Las Vegas. Cambridge University Press. Vizcaíno, S. Shelvocke, G. 1992a [1596] Relation of Sebastián Vizcaíno: August 1992 [1726] Cabo San Lucas: August 1721. In Ethnology of November 1596 (1990). Translated by M. Mathes. In the Baja California Indians, M. Mathes, ed., pp. 306-333. Ethnology of the Baja California Indians, M. Mathes, ed., [Spanish Borderland Sourcebooks 5.] New York: Garland pp. 135-144. [Spanish Borderland Sourcebooks 5.] New Publishing. York: Garland Publishing. Swadesh, M. 1992b [1602] Relation of Sebastián Vizcaíno: June-November 1967 Lexicostatistic Classification. In Handbook of Middle 1602 (1990). Translated by M. Mathes. In Ethnology of American Indians 5, R. Wachope, ed., pp. 79-115. Austin: the Baja California Indians, M. Mathes, ed., pp. 145-163. University of Texas Press. [Spanish Borderland Sourcebooks 5.] New York: Garland Troike, R. C Publishing. 1976 The Linguistic Classification of Cochimi. In Hokan Studies, M. Langdon and S. Silver, eds., pp. 159-63. The Vogt, W. P. Hague: Mouton. 2005 Dictionary of Statistics and Methodology: A Non

Turner, S., and R. Young technical Guide for the Social Sciences. 3rd Edition. 2007 Concealed Communities: The People at the Margins. Thousand Oaks: Sage Publications. International Journal of Historic Archaeology 11:297-303.

This content downloaded from 142.103.160.110 on Fri, 11 Jan 2019 05:39:20 UTC All use subject to https://about.jstor.org/terms